Corner Case Emerges in Jackpot Pick-6 Rules

[ScottJ]

For those not familiar with a "corner case", the term refers to a typically obscure situation in software engineering which might be a bug (or a feature), but requires an unusual set of circumstances to expose it.

Last Friday, a $432 ticket purchased on Del Mar's jackpot Pick-6 with a 20-cent minimum (2,160 combinations) was hit with a unique ticket resulting in a payout of $555,195.00. This was noteworthy not only because of the payoff, but we were once again chasing the definition of the word "unique".

Folks watching the Del Mar television screens that day might have seen the "wrong payoffs" posted as in fact there were believed to be two live tickets into the last leg.

How could that error be made? Well, here is the corner case.

In one of the sequence legs, a horse was scratched, moving that selection to the winning post time favorite, who was also on that same physical ticket. Therefore, the ticket effectively had the winning horse twice allowing some to argue that the winning ticket was no longer "unique".

However, since the ticket was purchased as one cohesive unit with a single serial number, the unique ticket criteria was met, hence the payout.

This might be the greatest case study for caveman ticket construction ever offered as it turns out to have been the saving grace for this player. If two different serial numbers had been generated, the bonus payout would have been voided.

[dilanesp]

Originally Posted by ScottJ

For those not familiar with a "corner case", the term refers to a typically obscure situation in software engineering which might be a bug (or a feature), but requires an unusual set of circumstances to expose it.

Last Friday, a $432 ticket purchased on Del Mar's jackpot Pick-6 with a 20-cent minimum (2,160 combinations) was hit with a unique ticket resulting in a payout of $555,195.00. This was noteworthy not only because of the payoff, but we were once again chasing the definition of the word "unique".

Folks watching the Del Mar television screens that day might have seen the "wrong payoffs" posted as in fact there were believed to be two live tickets into the last leg.

How could that error be made? Well, here is the corner case.

In one of the sequence legs, a horse was scratched, moving that selection to the winning post time favorite, who was also on that same physical ticket. Therefore, the ticket effectively had the winning horse twice allowing some to argue that the winning ticket was no longer "unique".

However, since the ticket was purchased as one cohesive unit with a single serial number, the unique ticket criteria was met, hence the payout.

This might be the greatest case study for caveman ticket construction ever offered as it turns out to have been the saving grace for this player. If two different serial numbers had been generated, the bonus payout would have been voided.

This indicates that the "fix" (which TBC was only a partial fix) they implemented is working- this guy deserved to get paid and was.

[BarchCapper]

Originally Posted by ScottJ

However, since the ticket was purchased as one cohesive unit with a single serial number, the unique ticket criteria was met, hence the payout.

Remember the Remington case involving the dead heat in their Jackpot Pick 6 from a couple of years ago where one guy had both horses on the same "unique" ticket? He did not get paid (at least not initially) - I could not find the final outcome. (Remington no longer offers a pick 6)

http://www.paceadvantage.com/forum/s...d.php?t=157448

[AskinHaskin]

Originally Posted by ScottJ

This might be the greatest case study for caveman ticket construction ever offered as it turns out to have been the saving grace for this player. If two different serial numbers had been generated, the bonus payout would have been voided.

The story was interesting, and the wager was offered and paid as intended, but there are more holes in the above paragraph than there ever were in the rules of the jackpot pick-6.


A - Serial numbers have zero to do with the Caveman Ticket concept. The cited bet was still a caveman wager to any extent which we know, regardless of how many hundreds or thousands of serial numbers were employed.


B - That same player could have had hundreds of different Serial numbers and still won the jackpot.


C - Nothing about the interesting story you shared endorses the caveman strategy at all.



RE:

A) If five humans decide to share a giant pick-6 wager entailing a single layout of combinations tallying $12,800 total, with most or all wagering online and splitting the cost @ $2560 each... IF they happen to go 5-deep in one leg, and for the purposes of simplicity, each "single" a different runner in that race so as to easily distribute the shared cost across wagering platforms, it is still a caveman ticket.


B) You cited "2160" combinations... Had the player somehow pinpointed a 40-cent wager/ticket with a single serial number including

W/W/W/S,W/W/W

it would not have mattered at all whether he had 1, or 2058 other serial numbers


C) The caveman strategy is perhaps fancied most for its simplicity... and it is poetic that those needing simplicity the most tend to have the largest wagering bankrolls, which match well to the caveman concept.

Anyone using only logical analysis without need or want for simplicity can determine that the caveman concept is far in arrears of weighted alternatives.

(of course there's always the guy who employed the caveman strategy to take down the only winning pick-5 on a big day at Belmont for upward of a quarter million, who will be a caveman until the day he dies)


I was glad to read the details you shared...

[ScottJ]

Originally Posted by AskinHaskin

B - That same player could have had hundreds of different Serial numbers and still won the jackpot.

RE:

B) You cited "2160" combinations... Had the player somehow pinpointed a 40-cent wager/ticket with a single serial number including

W/W/W/S,W/W/W

it would not have mattered at all whether he had 1, or 2058 other serial numbers

I believe we are saying the same thing, but for clarity, the following.

If the player purchased one single 20-cent ticket that was, using your notation, W/W/W/W/W/W, and then went back to the window and purchased another single 20-cent ticket that was W/W/W/S/W/W, these two tickets would have been assigned unique serial numbers (as separate bets) and the player would NOT have won the Jackpot.

Using an "A/B/C construction" technique can easily lead to this situation. In our example, the first ticket could have been all A selections while the second ticket was an A/A/A/B/A/A format.

The only way to have avoided this situation was to caveman all of the cross-products (using a set-theory term there) onto a single ticket with a single serial number.

[tbwinner]

Originally Posted by BarchCapper

Remember the Remington case involving the dead heat in their Jackpot Pick 6 from a couple of years ago where one guy had both horses on the same "unique" ticket? He did not get paid (at least not initially) - I could not find the final outcome. (Remington no longer offers a pick 6)

http://www.paceadvantage.com/forum/s...d.php?t=157448

This is exactly what I thought of when seeing this. Glad they have a fix.

I rarely if ever pay these jackpot wagers but occasionally will throw a few bucks if I have a leg or two with a longshot I like...I always double check to make sure I don't have 2 tickets in as of course it'd be my luck that I hit both and get a paltry payout.

[cj]

There are likely a lot more of these out there that may or may not turn up. I remember writing a program that worked perfectly for yours, then it crashed one day and it took me a while to figure out why.

I generated a worksheet for figure making that would only use the top 8 horses from each race. On this particular day, there was a dead heat for eighth. It has happened a few more times since and is now fixed, but you never know.

[Suff]

Originally Posted by AskinHaskin

The story was interesting,

Interesting and well told.